Pranav — oss/acc

hardcore ai / @Remotion / security
articulating agi internally /
building 24/7 @SlidesGPT

[Pranav — oss/acc @thecmdrunner]

511 days with these talented folks @Aeos_Labs,

✦ crazy mindset & insane growth
✦ a team that just “gets” it and has my back
✦ product ownership at a world class scale

[Varun Mayya @waiting4_asi]

Before I talk about Project 11 A (in a couple of days), we first need to talk about Aeos. Aeos is now 300+ full time employees, profitable, and bootstrapped. We’ve had a period of insane rapid growth lately.

Read what we did in 2024: https://t.co/1DvRgfP1OS

[Theo - t3.gg @theo]

I find myself giving this advice pretty often, figured I should share it publicly

Once you're paid enough to live, the ONLY thing that matters is that you're surrounded with awesome fucking people.

[Pranav — oss/acc @thecmdrunner]

@sutarabhiii @grok what if they’re stored as json 🙃

[Pranav — oss/acc @thecmdrunner]

asked Gemini's new decisions model to pick the best decisions API

it picked others 79/100 times 💀

[Google @Google]

We're introducing the Gemini Decisions API in @GoogleAIStudio, powered by our new Gemini Decision 1 model.

Instead of generating a text response token by token, Gemini Decision 1 evaluates your input (across text, images, audio, video, and PDFs) against up to 1,000 questions to make fast, confident decisions in a single pass, with zero output tokens billed.

Here's a look at how it works 🧵

[Pranav — oss/acc @thecmdrunner]

I compared this against Jev and OpenAI Decisions.

short answer: surprisingly good despite the spoiled reputation 🤔

[Google @Google]

We're introducing the Gemini Decisions API in @GoogleAIStudio, powered by our new Gemini Decision 1 model.

Instead of generating a text response token by token, Gemini Decision 1 evaluates your input (across text, images, audio, video, and PDFs) against up to 1,000 questions to make fast, confident decisions in a single pass, with zero output tokens billed.

Here's a look at how it works 🧵

[Pranav — oss/acc @thecmdrunner]

@Shokxye absolutely so true!

[Pranav — oss/acc @thecmdrunner]

@janwunschcom https://t.co/kk4S6fBb34

[Pranav — oss/acc @thecmdrunner]

@RipperdocMD @levelsio Stayed in a hotel, liked the room a lot and wished to copy some elements into mine.
Recorded a complete video, and then “make no mistakes” :)

[Pranav — oss/acc @thecmdrunner]

already making my ideal bedroom with claude 😄

[@levelsio @levelsio]

Claude Code can design furniture part by part and then give you a bill of materials to order each part and put it together yourself

And way cheaper than big interior brands sell you their furniture

Furnishing my house I discovered how absolutely insane prices are for the most basic cheap materials put together with a fancy brand tied to it

One shop had a sofa for $30,000 😂

I think with Opus 5.5 and other models easily operating 3d programs like Blender and FreeCAD (both free) it makes more sense to cut out the middle man and go straight to suppliers and design and build your own furniture

Which in a way makes customers more powerful (they can get exactly what they want and usually for cheaper without the markup)

Anyway @theannalux designed this table from scratch, she's not a furniture designer, not a coder, she just had an idea of a style and worked with Claude Code in the terminal (far superior to the app I will keep saying this)

If you design your own furniture and buy the raw materials directly, you can save 85% compared to premium brands like Poliform

That means the $30,000 sofa becomes $4,500! And the $8,000 dinner table becomes just $1,200!

You can see where the idea comes from that AI will make everything 5x to 20x cheaper, and it's already starting!

[Pranav — oss/acc @thecmdrunner]

@codeandfish @levelsio Chrome Plugin is all it needed, rest is @threejs

[Pranav — oss/acc @thecmdrunner]

@oKMz_ https://t.co/3mit02ImXk

[Pranav — oss/acc @thecmdrunner]

@RipperdocMD @levelsio Stayed in a hotel, liked the room a lot and wished to copy some elements into mine.
Recorded a complete video, and then “make no mistakes” :)

[Pranav — oss/acc @thecmdrunner]

@BrianVia @levelsio https://t.co/3mit02ImXk

[Pranav — oss/acc @thecmdrunner]

@RipperdocMD @levelsio Stayed in a hotel, liked the room a lot and wished to copy some elements into mine.
Recorded a complete video, and then “make no mistakes” :)

[Pranav — oss/acc @thecmdrunner]

@jishnusamal Chrome Plugin is all it needed, rest is @threejs

[Pranav — oss/acc @thecmdrunner]

@RipperdocMD @levelsio Stayed in a hotel, liked the room a lot and wished to copy some elements into mine.
Recorded a complete video, and then “make no mistakes” :)

[Pranav — oss/acc @thecmdrunner]

Punarv's claims are wrong.
Here' the full story:

LTV will be high: cuz your customers will live longer.

CAC will be too high: for hospitals, once everyone starts taking care of their health.

No one will pay for a health subscription: right, people rather pay the bigger bill later.

India is a trust deficit AND Vitamin D deficit market.

You're targeting only the 0.01% of India: that's 140K customers ready to pay, bigger TAM than most YC startups.

just keep going 🩵

[punarv @ycocerious]

It's just a blood test.
No one will pay for a subscription.
Design is not a moat.
Ultrahuman is going to kill you.
You don't have healthcare experience.
You need to find a co-founder.
You're not from an IIT.
You have no experience.
The CAC will be too high.
Orange Labs is going to kill you.
India is a trust deficit market.
Too early for preventative health in India.
Claude is going to kill you.
It's too competitive.
There are no margins in blood testing.
OpenAI is going to kill you.
You're not solving a big enough problem statement.
People will use AI to build their own software.
Blood tests are a commodity.
You don't have a moat.
It's too expensive.
You can't raise funds.
You should build your own wearable instead.
Google is going to kill you.
Indians don't care about brand.
You're targeting only the 0.01% of India.
You can't scale it with only organic marketing.
Your CAC to LTV ratio will be very low.
You should pivot.
You should be an AI health coach.
You can't build both product and distribution at the same time.
It's too risky.
Consumers won't pay for a premium product.
You should target tier 2 and tier 3.
Cent is going to kill you.
You aren't fast enough.
You're not seeing the big picture.
You're just a kid.
You're just an engineer, you need doctors.
...

just keep going.

Happy Diwali everyone 🪔🩵

[Pranav — oss/acc @thecmdrunner]

@ycocerious agency: $16k.
punarv: $16? k.

[Pranav — oss/acc @thecmdrunner]

openai patched 948 vulns in one run.
this is the new bar for agents.

Verifiable runs, with clear yes/no.
reproducible & dependable tests will shape everything.

[OpenAI @OpenAI]

Our security researchers built an autonomous cyber vulnerability discovery, exploitation, and patching pipeline.

In a single run with an upcoming cyber model, it validated and generated patches for 948 local privilege escalation vulnerabilities in the Linux kernel at a total API cost of $1.2 million.

[Pranav — oss/acc @thecmdrunner]

On my way to have S-tier Metabolism,
thanks to @ycocerious.

All the best with @nuehealth_!

[punarv @ycocerious]

This video marks our announcement to the world.

Nue Health is now open for public access.

This has been more than a year in the making, but we are now by far the best blood test in the market.

i promise you, we're just getting started.

[punarv @ycocerious]

This video marks our announcement to the world.

Nue Health is now open for public access.

This has been more than a year in the making, but we are now by far the best blood test in the market.

i promise you, we're just getting started.

[Pranav — oss/acc @thecmdrunner]

@ycocerious Only uphill from here.
Not just for a few, but for all those who refuse to die early!

[Pranav — oss/acc @thecmdrunner]

@ycocerious cant wait.